Kenny Workman’s Post

Are you at BioIT Boston next week and grappling with protein crystal detection in drug discovery? Our recent breakthrough, presented at the Center for Computational Biomedicine (CCB) seminar series at Harvard, offers a promising solution to this longstanding challenge. We managed to surpass the state of the art in the field, addressing the key problem of generalisation to new data sources. Our findings demonstrate: ✅ Performance: Achieving a crystal detection error rate reduction by 61%. ✅ With the capability to fine-tune to new data using as few as 60 images of crystals, our model showcases adaptability to diverse experimental setups. ✅ Computational Efficiency: By reducing computational effort by 98%, from 260 training epochs across 50 GPUs to just 3 epochs on a single GPU, we ensure cost-effective and scalable implementation. These advancements hold great potential for the pharmaceutical industry, promising substantial cost savings and accelerated timelines in drug development. We recognize the value of collaboration in driving innovation forward. If your organization possesses crystal images or datasets, we invite you to explore partnership opportunities. Together, we can leverage our cutting-edge technology to optimize crystal detection and propel drug discovery to new heights. Let's elevate drug discovery through collaboration and innovation. Reach out to initiate a dialogue and explore possibilities for mutual advancement. 🗓️ Interested in a meeting? Schedule here: https://lnkd.in/ghNiB3RW https://lnkd.in/dHkBE56h

Interesting, early #GenAI + #drugdiscovery read from June 2023; really, as we dig deeper, from April 2020 in the throes of the #covid19 pandemic. 😷 This study feeds IBM’s variational autoencoder CogMol with #opensource, curated datasets (ZINC, BindingDB) and analyzes outputs with GROMACS, RDKit, and RXN to analyze molecular props and assess #synthetic feasibility. Target: two regions of the emerging SARS-CoV-2 variants. 🦠 The overall result? “To our knowledge, the present study provides the first validated demonstration of a single generative machine intelligence framework that can propose previously unidentified and promising #inhibitors for different #protein drug #targets with a high success rate while only using protein #sequence information during design.” A sentence the authors liked so much, it appears in both the introduction and the discussion. 😉🧑🏻💻👩🏽💻 I applaud this broad-scale, academic + tech + industrial collaboration as the world was closing down. The 8 generated leads they produce do inhibit the SARS CoV-2 spike and main protease at uM concentrations. 🎉 However, these molecules are flat-land; nary a chiral center or conformational controller past some cycloalkanes. And the authors admit the model needs refinement in terms of an #ADME-tox (physical) screen, which would feed back and inform the Swiss-ADME predictions. Given the fast cycles and the world reopening, one imagines that further models (MolFormer?) are taking these parameters into account. 👨🏾🔬 For the paper: https://lnkd.in/efhJHQMN

AI and Medical Research: Can AI accelerate drug discovery? AI aids analysis, but the intricate process of drug development necessitates a collaborative approach involving biology, chemistry, and clinical research. 💊🔬🤖 #AIinDailyLife #SmartTechSolutions #AIAdvancements #TechIntegration #AIConvenience #FutureTechNow #EverydayAI #AIRevolution #AIAssistants #AIEnhancements

Language Models & Chemistry: Accelerating the Path to Novel Therapeutics! Combining LLMs with chemistry-specific tools, these intelligent systems are optimizing synthesis pathways & expediting drug discovery. Quick Read: https://lnkd.in/gCjNt4m9 #bioinformatics #cheminformatics #LLMs #chemistry #biochemistry #drugdiscovery #proteindesign #ai #machinelearning #sciencenews

Excited to share our latest study on P450 enzymes - a family known for their remarkable capabilities and complexities. Leveraging AI, we achieved a 4X increase in development speed. Dive into our findings here! #enzymes #syntheticbiology #ai #proteinengineering

From Biotech to “Tech-Bio” By Harper Chew Tech-bio is an exercise in applying bits onto atoms as technology seeks to manipulate and redefine biology through both hardware and software solutions, giving us a new understanding of biological data and techniques allowing for the design and construction of new biological parts, devices, and systems. This blogpost is the introduction to our two-part series on tech-bio. In this first part, we will be looking at the key innovations that enabled the shift from biotech to tech-bio, and how these can be applied in the biopharma world. #blogpost #techbio #biopharmaceuticals

We are proud to announce the results of a project with one of our earliest partners! 'The company' approached us with a project that they had made steady progress on, but not enough to continue on the same path. Using Cradle's AI-generated sequences, combined with the human expertise at 'the company', we were able to design P450 enzymes 4x faster! 'The company' set out on a quest to advance the biosynthesis of an active pharmaceutical ingredient. Their engineering goal was to increase the activity of a key P450 enzyme. Over 10 experimental rounds, bioengineers at the Company tested 1200 enzyme variants. The design of these variants combined a blend of cutting-edge bioinformatics, molecular modeling, and rational design techniques. To keep investing in enzyme engineering, bioengineers at 'the company' required a significant improvement in every round. Seeking to make the most of their project data, they partnered with Cradle to leverage our custom-trained generative Machine Learning models. After 'the company' initially completed ten rounds of mutagenesis, showcasing steady progress, Cradle's technology elevated the game by boosting substrate conversion (nearly tripling in 3 cycles) with a 4X rate of improvement over time. “Cradle's AI-based approaches were invaluable in enhancing our enzyme's activity. We were particularly impressed by its ability to effectively leverage our historical data, as well as the collaborative spirit the Cradle team exhibited throughout the project.” Read more: https://lnkd.in/ejNjbpTY

Mark your calendars for June 27th to attend our upcoming webinar : "An Introduction to Computational Cytometry for Drug Discovery". See you there! #Webinar #ComputationalCytometry #DrugDiscovery 🌐 Click on the link 👇 to learn more!

Biotech & Pharma Companies Are Super Excited About AI ML & Bioinformatics - Here's Why! #Biotech #Pharma #AI #MachineLearning #Bioinformatics #IndustryTrends #Innovation #Tech #AITech #MLTech #BiotechTrends #PharmaTrends #Trending

Learn how Diamond Age transformed a pharma company's bioinformatics pipeline, reducing processing time from 19 hours to just 1.5 hours and cutting compute costs by 90%. This optimization boosts research efficiency and accelerates decision-making, driving innovation forward. https://lnkd.in/d2g3TTKj For inquiries on optimizing your computational pipeline, contact us at [email protected] or visit https://lnkd.in/eWCymszj. #Bioinformatics #PharmaEfficiency #HighThroughputSequencing #PipelineOptimization #CloudComputing #DrugDiscovery #DiamondAge